CryptNoSQL – A Methodology for Secure Querying and Processing of Encrypted NoSQL Data on the Cloud Environment

Abstract

Businesses today use modern computing technologies such as Big Data and Machine Learning in their daily operations, which require effective management of large amounts of data. Relational data formats are no longer suitable for these applications, and NoSQL data formats are preferred. The use of cloud infrastructure offers advantages such as scalability, availability, and resource maintenance, but data security remains a challenge. Although cloud vendors provide encryption features, they may not be sufficient for sensitive data. To address this, some businesses use their encryption methods, but retrieving data from an encrypted form may not be possible. While specific encryption methods support the processing of encrypted data without decryption, there is no complete implementation of secure processing for NoSQL data from MongoDB or other databases. The proposed methodology, called CryptNoSQL, provides a secure way to query and process NoSQL data, including updates on encrypted data. We introduce a customized database design model that selects an appropriate encryption method during the insertion of a document based on the type of field and the operation it will be involved in. Our experimental results demonstrate that our approach is suitable for organizations with sensitive data hosted on the cloud and that require frequent query operations on this data.